Dear All,
I would like to ask how I should visualize the mRNA expression of a single gene (HNF1A) on a graph. I have my positive control line (HepG2), in which the expression of the gene of interest is definitely occurring, and three different hiPSC-ECs lines (one control and two mutants) in which I would like to check the expression of HNF1A (between hiPCS-EC lines; as well as hiPSC-EC lines and HepG2).
I noticed that the Ct of my reference gene (EF2) is different in hiPSC-ECs and HepG2, so I cannot use 2^(delta delta Ct). Can I just show the differences in the Ct of my gene of interest and use such values for statistics?
Thank you in advance,
Dawid
Dear colleague,
the best and most accepted way of working with gene expression is using relative expression, based on a reference gene (housekeeping gene).
It is not recommended to use only the Ct value of the target gene in comparisons between different "treatments".
Other normalization methods can be used (doi: 10.1038/sj.gene.6364190.), however the 2^deta Ct or the 2^(delta delta Ct) is best.
If you have already evaluated the stability of the reference gene, you could try to amplify another gene as a reference gene.
If it is not possible, the suggestion is to do a test with reference gene already amplified to verify its stability, sometimes, in numerical terms, the Cts appear different, but under a test, it is concluded that the gene is stable (See the methods used in these works: https://doi.org/10.1371/journal.pone.0204348 and https://doi.org/10.1371/journal.pone.0127935).
Another strategy could be to use a more robust model, such as a mixed model considering the cell culture sample as a random effect (doi: 10.1016/j.ygeno.2009.04.008).
But, by no means, it is recommended to use only the Ct to test or to present the results.
Hope this help you.
Good luck.
Thank you Margareth E. Botelho! To sum up, you recommend finding a better reference gene or using the mixed model, am I right?
Dawid
In my experience, consecutively expressed RNA is a better choice for semi-quantifying RNA expression especially you need to perform cross-cell-line comparison. I usually use 18S rRNA as my internal control of RT-PCR which expression across cell lines are comparable.
Can you mention the reference gene you used? I prefer RP49 or 18s RNA depending on how you synthesized your cDNA (if using oligodT kit, 18s RNA cannot be used). If you are getting very different Ct values for your reference genes, check if you are starting with equal amount of RNA. If your initial RNA is different, Ct values will vary. Test 2-3 reference genes and decided which works for your experiments. For cancer samples, many 'houskeeping' genes like actin/gapdh do not work .
Bhagyashree Kaduskar I work with endothelial cells and I use EF2 as a housekeeping gene.
Dawid, is this reported to work best as housekeeping gene? I would suggest using some robust ribosomal gene if possible.
- Badges
- Science method